Identification of texture MRI brain abnormalities on first-episode psychosis and clinical high-risk subjects using explainable artificial intelligence

Κορδά, Αλεξάνδρα; Andreou, Christina; Rogg, Helena; Avram, Mihai; Ruef, Anne; Davatzikos, Christos; Koutsouleris, Nikolaos; Borgwardt, Stefan

doi:10.1038/s41398-022-02242-z

Cited by 12 publications

(5 citation statements)

References 103 publications

(129 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several diverse fields have embraced the explainable component of AI, prioritising trustworthiness over accuracy. XAI has been applied in drug discovery [99,100], industrial applications [101,102], gaming [103,104], neurological disorders [105,106,107], neuroscience [108,109] and recommender systems [110,111]. This tremendous growth has led to several XAI-based review articles in the healthcare domain in the past years.…”

Section: Introductionmentioning

confidence: 99%

Application of Explainable Artificial Intelligence in Alzheimer's Disease Classification: A Systematic Review

Vimbi¹,

Shaffi²,

Mahmud

et al. 2023

Preprint

View full text Add to dashboard Cite

Context: Artificial Intelligence (AI) in the medical domain has achieved remarkable results on various metrics primarily due to recent advancements in computational capabilities. However, AI models used for Computer Aided Diagnosis (CAD) have limited acceptance and trust due to their typical {Blackbox} nature. Consequently, explainable AI (XAI) is needed to justify the trustworthiness of the machine's predictions. Approach: In this work, we undertake a systematic review based on PRISMA guidelines on research articles investigating Alzheimer's Disease (AD) prediction with XAI for the past decade. The review was driven by carefully formulated research questions (RQ) which helped categorize AI models into different XAI methods(Post-hoc, Ante-hoc, Model-Agnostic, Model-Specific, etc) and frameworks (LIME, SHAP, GradCAM, LRP, etc.). Summary: This categorisation provides broad coverage of the interpretation spectrum from intrinsic (model-specific, ante-hoc models) to complex patterns (model-agnostic, post-hoc models) and by taking local explanations to a global scope. Additionally, the study categorizes merit in terms of different forms of interpretation to provide in-depth insight into the factors that support the clinical diagnosis of AD. Finally, limitations and open research challenges are outlined, and possible prospects are presented.

show abstract

Section: Introductionmentioning

confidence: 99%

Application of Explainable Artificial Intelligence in Alzheimer's Disease Classification: A Systematic Review

Vimbi¹,

Shaffi²,

Mahmud

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…With the availability of post hoc explainable methods [31], explaining black-box DL models has become easier. Studies such as [30,[32][33][34] use such explainability techniques on ML models. Nevertheless, the use of explainability has been limited to either the extraction of the most influential model features/inputs or providing model-specific improvements to outputs.…”

Section: Introductionmentioning

confidence: 99%

Towards Personalised Depression Modelling and Explanation from Wearable Data

Chatterjee

Mishra²,

Sundram

et al. 2023

Preprint

View full text Add to dashboard Cite

Depression and anxiety are the leading causes of health loss globally, and the Covid-19 pandemic has significantly exacerbated the effect of these disorders. There is a widening gap between available resources and mental health needs globally. Digital health applications using artificial Intelligence (AI) are a promising opportunity to address this widening gap. Increasingly, passively acquired data from wearables is augmented with carefully selected active data from the participants to develop machine learning (ML) models of depression. However, these ML models are black-box in nature, and hence the outputs are not explainable. Depression is also multi-modal, and the reasons for depression may vary significantly between individuals. Explainable and personalised models will thus be beneficial to clinicians for determining the main features that lead to a decline in the mood state of a patient, thus enabling suitable personalised therapy. This is currently lacking. Therefore, this study presents the first methodology for developing personalised and accurate deep learning (DL)-based models for depression, along with novel methods for identifying the key facets that lead to the exacerbation of depressive symptoms. We illustrate our approach using an existing multi-modal dataset containing longitudinal ecological momentary assessments of depression, lifestyle data from wearables, and neurocognitive assessments for 14 mild to moderately depressed participants over one month. We train classification- and regression-based DL models to predict participants’ mood scores - a discrete score given to a participant based on the severity of their depressive symptoms. The models are trained inside eight different evolutionaryalgorithm-based optimisation schemes that optimise the model parameters for maximum predictive performance. A 5-fold cross-validation scheme is used to verify the model performance, with model error as low as 6% for some participants. We use the best model from the optimisation process to extract indicators, using SHAP, ALE and Anchors from Explainable AI literature to explain why certain predictions are made and how they affect mood. These feature insights can assist health professionals in incorporating personalised interventions into the patient’s treatment regimen.

show abstract

“…Studies such as [ 43 , 44 , 45 ] use explainability techniques on ML models to obtain insights into the model outputs. Moreover, recent works have begun exploring explainability in mental health settings [ 24 , 46 , 47 , 48 , 49 ].…”

Section: Introductionmentioning

confidence: 99%

Towards Personalised Mood Prediction and Explanation for Depression from Biophysical Data

Chatterjee,

Mishra,

Sundram

et al. 2023

Sensors

View full text Add to dashboard Cite

Digital health applications using Artificial Intelligence (AI) are a promising opportunity to address the widening gap between available resources and mental health needs globally. Increasingly, passively acquired data from wearables are augmented with carefully selected active data from depressed individuals to develop Machine Learning (ML) models of depression based on mood scores. However, most ML models are black box in nature, and hence the outputs are not explainable. Depression is also multimodal, and the reasons for depression may vary significantly between individuals. Explainable and personalised models will thus be beneficial to clinicians to determine the main features that lead to a decline in the mood state of a depressed individual, thus enabling suitable personalised therapy. This is currently lacking. Therefore, this study presents a methodology for developing personalised and accurate Deep Learning (DL)-based predictive mood models for depression, along with novel methods for identifying the key facets that lead to the exacerbation of depressive symptoms. We illustrate our approach by using an existing multimodal dataset containing longitudinal Ecological Momentary Assessments of depression, lifestyle data from wearables and neurocognitive assessments for 14 mild to moderately depressed participants over one month. We develop classification- and regression-based DL models to predict participants’ current mood scores—a discrete score given to a participant based on the severity of their depressive symptoms. The models are trained inside eight different evolutionary-algorithm-based optimisation schemes that optimise the model parameters for a maximum predictive performance. A five-fold cross-validation scheme is used to verify the DL model’s predictive performance against 10 classical ML-based models, with a model error as low as 6% for some participants. We use the best model from the optimisation process to extract indicators, using SHAP, ALE and Anchors from explainable AI literature to explain why certain predictions are made and how they affect mood. These feature insights can assist health professionals in incorporating personalised interventions into a depressed individual’s treatment regimen.

show abstract

Identification of texture MRI brain abnormalities on first-episode psychosis and clinical high-risk subjects using explainable artificial intelligence

Cited by 12 publications

References 103 publications

Application of Explainable Artificial Intelligence in Alzheimer's Disease Classification: A Systematic Review

Application of Explainable Artificial Intelligence in Alzheimer's Disease Classification: A Systematic Review

Towards Personalised Depression Modelling and Explanation from Wearable Data

Towards Personalised Mood Prediction and Explanation for Depression from Biophysical Data

Contact Info

Product

Resources

About